By using the panel unit root, co-integration test and panel-based error correction models, according to provincial panel data of 30 provinces and municipalities of China during 1978—2007, this paper makes empirical studies on the relationship between agricultural mechanization and peasants’ income, the results show that there is long-run equilibrium relationship between agricultural technical progress and peasants’ income, that from short-term view, there is unidirectional causality from peasants’ income to agricultural mechanization in the east region, however, there is no any causality in any direction in the middle and west region, and that from longterm view, there is bidirectional causality between agricultural mechanization and peasants’ income in the east and west region, however, there is unidirectional causality from agricultural mechanization to peasants’ income in the middle region. When making relative policies, Chinese Government should start with technical progress to highlight long-term effect of peasants’ income growth and comprehensively consider different causalities in different regions between agricultural mechanization and peasants’ income.